Your next step
Language models have different designs
Encoder, decoder, encoder-decoder, embeddings, and the alternatives.
Start learningWhat you’ll learn
Encoder-only models represent an input
Recognize an encoder-style model used to classify or represent supplied text.
About 4 minutes · Open activity
Masked practice can use both sides
Identify which context supports a masked-text objective.
About 4 minutes · Open activity
Decoder-only models extend a sequence
Recognize a GPT-style causal generation route.
About 4 minutes · Open activity
Encoder-decoder models transform one sequence into another
Recognize a T5-style source-to-output design.
About 4 minutes · Open activity
Embedding models return representations
Distinguish a retrieval representation from generated prose.
About 4 minutes · Open activity
Broad pretraining can support many tasks
Explain a foundation model as a reusable starting point.
About 4 minutes · Open activity
Transformers are not the only sequence design
Recognize recurrent, state-space and hybrid alternatives at a high level.
About 4 minutes · Open activity
Read three fictional model cards
Match a representation task, a continuation task and a source-transformation task to compatible designs, and explain why architecture alone does not establish quality.
About 6 minutes · Open activity